evaluation of nine commercial sars-cov-2 immunoassays · 4/9/2020 · page 1 of 15 evaluation of...
TRANSCRIPT
Page 1 of 15
Evaluation of nine commercial SARS-CoV-2 immunoassays
Ria Lassaunière1, Anders Frische1, Zitta B. Harboe2,3, Alex C.Y. Nielsen4, Anders Fomsgaard1, Karen A. Krogfelt1,5, Charlotte S. Jørgensen1*
1Department of Virus & Microbiological Special Diagnostics, Statens Serum Institut, Denmark 2Department of Infectious Diseases and Pulmonary Medicine, Nordsjællands Hospital, Copenhagen University Hospital, Denmark
3Department of Bacteria, Parasites and Fungi, Statens Serum Institut, Copenhagen, Denmark 4Department of Clinical Microbiology, Rigshospitalet, Copenhagen, Denmark 5Department of Molecular and Medicinal Biology, Institute for Science and Environment, Roskilde University, Denmark
*Corresponding author: E-mail: [email protected], phone: +45 3268 8248
Abstract
Due to urgency and demand, numerous severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)
immunoassays are rapidly being developed and placed on the market with limited validation on clinical
samples. Thorough validation of serological tests are required to facilitate their use in the accurate diagnosis
of SARS-CoV-2 infection, confirmation of molecular results, contact tracing, and epidemiological studies. This
study evaluated the sensitivity and specificity of nine commercially available serological tests. These included
three enzyme-linked immunosorbent assays (ELISAs) and six point-of-care (POC) lateral flow tests. The assays
were validated using serum samples from: i) SARS-CoV-2 PCR-positive patients with a documented first day
of disease; ii) archived sera obtained from healthy individuals before the emergence of SARS-CoV-2 in China;
iii) sera from patients with acute viral respiratory tract infections caused by other coronaviruses or non-
coronaviruses; and iv) sera from patients positive for dengue virus, cytomegalovirus and Epstein Barr virus.
The results showed 100% specificity for the Wantai SARS-CoV-2 Total Antibody ELISA, 93% for the Euroimmun
IgA ELISA, and 96% for the Euroimmun IgG ELISA with sensitivities of 90%, 90%, and 65%, respectively. The
overall performance of the POC tests according to manufacturer were in the rank order of AutoBio
Diagnostics > Dynamiker Biotechnology = CTK Biotech > Artron Laboratories > Acro Biotech ≥ Hangzhou
Alltest Biotech. Overall, these findings will facilitate selection of serological assays for the detection SARS-
CoV-2-specific antibodies towards diagnosis as well as sero-epidemiological and vaccine development
studies.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 2 of 15
Introduction
In December 2019, a novel coronavirus causing severe acute respiratory symptoms emerged in Wuhan, China
[1]. The World Health Organization (WHO) termed the disease, coronavirus disease 2019 (COVID-19), and
the causative virus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). As of 7 April, the virus has
spread to 212 countries and territories with 1 279 722 confirmed cases and 72 614 deaths worldwide [2]. At
present, the epidemic within the majority of countries have not yet reached its peak with the number of
cases and deaths predicted to rise in the coming weeks and months.
Accurate diagnosis of COVID-19 is essential, not only to ensure appropriate patient care but also to facilitate
identification of SARS-CoV-2 infected people, including asymptomatic carriers, who need to be isolated to
limit virus spread. The WHO recommends nucleic acid detection of SARS-CoV-2 in respiratory samples for the
diagnosis of COVID-19. Unfortunately, in the face of the rapidly growing epidemics worldwide, an increased
demand for diagnostic tests has led to a critical shortage in operational material for respiratory sample
collection and within the molecular diagnostic workflow [3,4]. This impedes rapid large scale testing, a
necessity for controlling the epidemic. Moreover, the heterogeneity of respiratory sample material and
anatomical location of sample collection, for example throat swab, saliva or endotracheal aspirate, affect the
sensitivity of SARS-CoV-2 viral nucleic acid testing [5,6]. Overall, there is an urgent need to identify alternative
diagnostic means.
Antibody testing, either using enzyme-linked immunosorbent assay (ELISA) or point-of-care (POC) lateral flow
immunoassays, may overcome some of these challenges. SARS-CoV-2-specific antibodies can be detected in
in serum of approximately 40% of COVID-19 patients as early as seven days after the onset of symptoms,
with seroconversion rates rapidly increasing to >90% by day 14 [7]. In recent studies, antibody testing has
been shown to be more sensitive than viral nucleic acid detection after approximately eight days of COVID-
19 illness duration [7,8]. While the combination of PCR and antibody tests is optimal for accurate diagnosis
[6], antibody detection will be particularly relevant for the later stages of infection where the virus has been
eliminated [5]. In addition to the diagnostic value of antibody testing, it will identify individuals who
developed immunity after infection that may protect against subsequent re-infection [9], as well as define
and monitor the extent of virus spread and a population’s herd immunity on a societal level.
Antibody testing may therefore be relevant in the following settings: i) diagnosis of patients who seek medical
attention more than seven after the onset of symptoms; ii) contact tracing; iii) determining potential
immunity and risk of infection; and iv) sero-epidemiological studies to understand the extent of COVID-19
spread. Due to urgency and demand, a lot of serological tests are rapidly being developed and made available
on the market with only limited validation on clinical samples. To address this, the present study evaluated
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 3 of 15
three ELISA assays and six POC lateral flow immunoassays for the detection of SARS-CoV-2 antibodies in
patients with COVID-19. CE-marked SARS-CoV-2 ELISA assays and POC tests were selected based on
availability in Denmark at the time of testing.
Materials & Methods
Study design
Retrospective study evaluating the sensitivity and specificity of commercially available immunoassays for the
detection of antibodies specific to the SARS-CoV-2 virus. Patient serum samples used in this study were
submitted to the routine serology laboratory at Statens Serum Institut for diagnostic purposes.
Serum samples
Case serum samples were obtained from COVID-19 patients (n = 30) admitted to the intensive care unit at
Hillerød Hospital, Denmark. SARS-CoV-2 infection was confirmed by viral nucleic acid detection in samples
from the respiratory tract. Control serum samples included archived anonymous serum samples obtained
from healthy blood donors 18-64 years with no history of SARS-CoV-2 infection (n = 10) and no recent travel
history, sera from patients with acute viral respiratory tract infections caused by other coronaviruses (n = 5)
or non-coronaviruses (n = 45), and sera from patients positive for dengue virus (n = 9), cytomegalovirus (CMV;
n = 2) and Epstein Barr virus (EBV; n = 10). One patient was positive for both CMV and EBV.
ELISA assays
The Wantai SARS-CoV-2 Ab ELISA (Beijing Wantai Biological Pharmacy Enterprise, Beijing, China; Cat # WS-
1096) was performed according to the manufacturer’s instructions. The assay is based on a double-antigen
sandwich principle that detects total antibodies binding SARS-CoV-2 spike protein receptor binding domain
(RBD) in human serum or plasma. Briefly, 100 μl undiluted serum samples were added to wells coated with
recombinant SARS-CoV-2 antigen and incubated for 30 minutes at 37 °C. Wells were washed five times
followed by the addition of HRP-conjugated SARS-CoV-2 antigen and subsequent incubation for 30 minutes
at 37 °C. Wells were washed five times and a chromogen solution was added. Following 15 minutes of
incubation at 37 °C, the reaction was stopped and the resultant absorbance was read on a microplate reader
at 450 nm with reference at 620 nm. The cut-off value for a positive result was calculated according to the
manufacturer’s instruction by adding the calculated negative control value to 0.160.
The Anti-SARS-CoV-2 IgG and IgA ELISAs (Euroimmun Medizinische Labordiagnostika, Lübeck, Germany; Cat
# EI 2668-9601 G and EI 2606-9601 A, respectively) were performed according to the manufacturer’s
instructions. In two separate semi-quantitative ELISAs, either IgA or IgG antibodies against SARS-CoV-2 spike
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 4 of 15
protein subunit 1 (S1) are detected in human serum or plasma. Briefly, 1:101 diluted serum samples were
added to wells coated with recombinant SARS-CoV-2 antigen and incubated for 60 minutes at 37 °C. Wells
were washed three times followed by the addition of HRP-conjugated anti-human IgA or IgG and subsequent
incubation for 30 minutes at 37 °C. Wells were washed three times and a chromogen solution was added.
Following 30 minutes of incubation at room temperature, the reaction was stopped and the resultant
absorbance was read on a microplate reader at 450 nm with reference at 620 nm. A ratio between the
extinction of the sample and calibrator on each plate were calculated. According to the manufacturer’s
recommendations, a ratio <0.8 is considered negative, ≥0.8 and <1.1 borderline, and ≥1.1 positive. However,
for sensitivity and specificity, 1.1 was used as a more stringent cut-off value for positive results and all values
<1.1 were considered negative.
Point-of-care (POC) tests
Six POC tests for rapid detection of antibodies in blood, serum or plasma were evaluated: 2019-nCOV IgG/IgM
Rapid Test (Dynamiker Biotechnology, Tianjin, China Cat # DNK-1419-1), OnSiteTM COVID-19 IgG/IgM Rapid
Test (CTK Biotech, Poway, CA, USA; Cat # R0180C), Anti-SARS-CoV-2 Rapid Test (AutoBio Diagnostics,
Zhengzhou, China; Cat # RTA0204), Coronavirus Diseases 2019 (COVID-19) IgM/IgG Antibody Test (Artron
Laboratories, Burnaby, Canada; Cat # A03-51-322), 2019-nCoV IgG/IgM Rapid Test Cassette (Acro Biotech,
Rancho Cucamonga, CA, USA; Cat # INCP-402), and 2019-nCoV IgG/IgM Rapid Test Cassette (Hangzhou Alltest
Biotech, Hangzhou, China; Cat # INCP-402). The tests were performed at room temperature according to the
manufacturer’s instructions. For all tests, the recommended sample volume of 10 μl serum was added to the
specimen well on the individual test cassettes followed by the addition of the supplied buffer. The buffer
volume differed by manufacturer and was added accordingly (60 μl, three drops, two drops, two drops, two
drops and 60 μl, respectively). The result was read visually after 10 minutes. Weak signals for IgM and IgG,
together or separate, was considered positive.
Statistical analyses
Sensitivity was defined as the proportion of patients correctly identified as having SARS-CoV-2 infections, as
initially diagnosed using nucleic acid detection of SARS-CoV-2 in respiratory samples. Specificity was defined
as the proportion of SARS-CoV-2 immune naïve study participants accurately identified as negative for
COVID-19. The clinical accuracies of the ELISA assays were examined by using Receiver Operator
Characteristic (ROC) plots with GraphPad Prism version 8.0.2 (GraphPad Software, San Diego, CA, USA). ROC
area under the curve (AUC) were calculated as the fraction “correctly identified to be positive” and the
fraction “falsely identified to be positive” determined according to manufacturer cut-off values for positive
results.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 5 of 15
Results
Sensitivities and specificities of the ELISA assays
Three commercial CE-marked ELISA assays for detecting SARS-CoV-2 antibodies were evaluated using 30
serum samples from PCR-positive cases with SARS-CoV-2 and 82 control serum samples. Twenty nine of the
30 cases (97%) were positive for SARS-CoV-2-specific antibodies by at least one of the three ELISA assays. In
one case, only a positive IgA result was detected, while in another case antibody responses were negative by
all tests.
The distance of data points from the manufacturer recommended cut-off values and confidence in assigning
a positive or negative status differed between the three assays (Figure 1A). The distribution of positive and
negative data points were distinct for the Wantai Total Ab assay, with a cut-off value above all the control
sera samples, which allowed for unequivocal interpretation. Conversely, the Euroimmun IgA and IgG assays
data had a less distinct separation, resulting in a ‘grey zone’ of borderline data points to which a positive or
negative status could not be assigned. Both case and control sera had borderline or inconclusive data points.
The sensitivities and specificities are shown in table 1. The sensitivity of the Wantai Total Ab ELISA was
equivalent to that of Euroimmun’s IgA ELISA at 93% and was greater than that observed for Euroimmun’s IgG
ELISA at 67%. The specificity of the Wantai Total Ab ELISA was 100% compared to 93% and 96% for the
Euroimmun IgA and IgG ELISAs, respectively. The positive predictive value and negative predictive value was
the highest for the Wantai Total Ab ELISA at 100% and 98%, respectively, compared to the Euroimmun IgA
ELISA (82% and 97%, respectively) and IgG ELISA (87% and 89%, respectively). The Euroimmun IgA ELISA
cross-reacted primarily with serum that contained antibodies to more than one respiratory virus (4/6 [67%])
and associated with the presence of adenovirus antibodies (5/6 [83%]) and dengue virus antibodies (Table
2). The Euroimmun IgG ELISA cross-reacted with a serum sample positive for human coronavirus HKU1 and
two samples with adenovirus antibodies.
ROC analysis
ROC AUC analysis presents a good parameter for the diagnostic power of an individual test and were
compared between the different ELISA kits (Figure 1B). The Wantai total antibody kit had the highest measure
at 0.973 (95% CI: 0.921-1.000), followed by Euroimmun’s IgA ELISA with 0.954 (95% CI: 0.897-1.00) and
Euroimmun’s IgG ELISA with 0.887 (95% CI: 0.810-0.964).
Sensitivities and specificities of the POC tests
Four POC tests were tested on all 30 case serum samples and had sensitivities in the rank order of 93% for
AutoBio Diagnostics, 90% Dynamiker Biotechnology and CTK Biotech, and 83% for Artron Laboratories (Table
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 6 of 15
1). The positive predictive value of these tests were 100%, while the negative predictive values were 91%,
89%, 89%, and 74%, respectively. The Acro Biotech was evaluated on five case serum samples and had a
specificity of 80%. One case serum sample was tested with the Hangzhou AllTest Biotech test and was positive
for both IgM and IgG.
The specificity of the six POC tests were evaluated primarily on control samples that showed some cross-
reactivity in the SARS-CoV-2 ELISA assays; the number of control sera tested varied between the different
POC tests (Table 1). The POC tests manufactured by Dynamiker Biotechnology, CTK Biotech, AutoBio
Diagnostics and Artron Laboratories had a 100% specificity, whereas the test from Acro Biotech and
Hangzhou AllTest Biotech had a specificity of 80% and 87%, respectively. For the latter two tests, cross-
reactivity was only observed for IgM. The Acro Biotech test cross-reacted with a control serum sample from
a human coronavirus HKU1 patient.
Antibody detection relative to duration of illness
To evaluate the sensitivities of the assays at different stages of COVID-19 disease, case sera were grouped
according the duration of disease: early phase, 7 to 13 days after the onset of disease symptoms; middle
phase, 14 to 20 days after the onset of disease symptoms; and late phase, ≥21 days after the onset of disease
symptoms. The sensitivities of the assays ranged from 40 to 86% for the early phase samples, 67 to 100% for
the middle phase samples, and 78 to 89% for the late phase (Figure 2).
In the early phase, the Wantai Total Ab ELISA had a sensitivity of 71% that plateaued at 100% after 10 days
of illness duration. The IgG ELISA had the lowest sensitivity at all three phases that showed a distinct increase
with each consecutive phase i.e. 43% in the early phase, 67% in the middle phase, and 78% in the late phase.
While the four POC tests evaluated according to illness duration were often weakly positive or detected only
IgG or IgM during the early phase (data not shown), their sensitivities were comparable to the Wantai Total
Ab ELISA and Euroimmun IgA ELISA in all three phases. In the early phase, a case sample that was negative
by both Total Ab and IgG was positive in the IgA ELISA.
Agreement between serological assays
To determine the agreement between the different ELISAs and POC tests evaluate, the proportion of case
sera that shared the same result between two assays were calculated. Despite comparable sensitivities of
certain assays, the tests did not necessarily give the same result in all instances (Figure 3A). The only tests
that were 100% concordant were the Dynamiker Biotechnology and CTK Biotech POC tests (Figure 3B). The
Wantai Total Ab ELISA and Euroimmune IgA ELISA was 93% concordant, whereas the highest agreement
between an ELISA and POC test was 93% between the Wantai Total Ab ELISA and AutoBio Diagnostics POC
test.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 7 of 15
Discussion
In the present study, three SARS-CoV-2-specific commercial ELISA assays and six POC rapid tests were
evaluated using sera from hospitalized adult patients with PCR-confirmed diagnoses for SARS-CoV-2 and a
collection of control serum samples taken before the emergence of the virus in China in December 2019.
Overall, the Wantai Total Ab ELISA had superior sensitivity and specificity compared to both Euroimmun IgA
and IgG ELISAs. The POC tests varied notably, with the best performance observed for the test produced by
AutoBio Diagnostics, followed by the tests produced by Dynamiker Biotechnology and CTK Biotech.
The differences observed for the sensitivity and specificity of the SARS-CoV-2-specific total antibody testing
and antibody type ELISAs correspond to previous reports. The Wantai Total Ab ELISA performed as reported
by the manufacturer (94.5% sensitivity and 100% specificity) and a separate study (93.1% sensitivity) where
the Total Ab ELISA, IgM ELISA and IgG ELISA produced by Beijing Wantai Biological Pharmacy Enterprise were
compared [7]. In the latter study, the IgM and IgG ELISAs had lower sensitivities (83% and 65%, respectively)
compared to the Total Ab ELISA (93.1%) [7]. This notably lower sensitivity for SARS-CoV-2-specific IgG
detection is in agreement to that observed here for the Euroimmun IgG ELISA (65%). In the present study the
Wantai IgG and Euroimmun IgG ELISAs could not be compared due to unavailability of the former. The
possibility that overall lower sensitivity of SARS-CoV-2 IgG ELISAs may be a more universal occurrence rather
than manufacturer dependent warrants further investigation. In addition to lower sensitivities, the
Euroimmun IgA and IgG ELISAs are also more prone to cross-react with negative sera as described in the
present study and in a separate analysis of the beta-versions of these assays [10].
The differences between the assays may, in part, be explained by the SARS-CoV-2 antigen targeted and the
ELISA format used. Both kits detect antibodies to the S1 subunit of the SARS-CoV-2 spike protein; however,
the Wantai Total Ab ELISA only targets the RBD within the S1. The RBD represents approximately 33% of the
S1 subunit, thus epitopes that may be recognized by cross-reacting epitopes outside of this domain are
absent. Furthermore, the RBD is highly diverse between SARS-CoV-2 and other beta-coronaviruses (hCoV-
OC43 and hCoV-HKU1), which may further reduce the likelihood of cross-reaction with these circulating
coronaviruses. Moreover, the Wantai Total Ab ELISA uses an antigen-antibody-antigen(peroxidase) format
whereas the Euroimmun ELISAs employ an antigen-antibody-antibody(peroxidase) format. The specificity of
the former is determined by a single antibody, whereas the latter has a second antibody that may introduce
additional specificities. The antigen-antibody-antibody format is required to distinguish between specific
antibody types, but may not necessarily have lead to decreased specificity as shown for in-house ELISAs [10].
The clinical sensitivity of IgM for early diagnosis of COVID-19 is currently unclear. SARS-CoV-2-specific IgM
does not consistently appear before its IgG counterpart, with some studies reporting detection of SARS-CoV-
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 8 of 15
2 spike protein-specific IgG before IgM [6,7,11]. While all the POC tests evaluated in this study are capable of
detecting both SARS-CoV-2 IgM and IgG antibodies, the majority detected both antibody types
simultaneously, even in the early convalescent phase, while some detected only IgG and others only IgM
(data not shown). On the contrary, cross-reactive IgM antibodies resulted in decreased specificity of two POC
tests evaluated. This IgM-associated reduced specificity was not observed for all POC tests, thus other
unknown factors define POC test specificity. POC tests are, by definition, often performed beside the patient,
whereas ELISAs are conducted in laboratories. This study demonstrated that while certain ELISA assays and
POC tests may share similar sensitivities and specificities, their results may not necessarily correspond for a
single patient sample. These discrepancies should be noted where SARS-CoV-2 ELISAs are used to confirm
POC test results.
Since the appearance of antibodies is time dependent, diagnosis of COVID-19 by serological methods is
limited to patients with a longer duration of illness. Within seven days of symptom onset or in the acute
phase of disease, nucleic acid detection of SARS-CoV-2 in respiratory samples is superior to antibody
detection for the diagnosis of COVID-19 [7,8,12]. However, after eight days of illness, the sensitivity of
serological assays surpasses that of nucleic acid testing [7,8]. Here we reported a 100% seropositivity in
patients 10 days after the onset of symptoms. It is unclear whether the latter sensitivity can be extrapolated
to mild COVID-19 cases, since the present study comprised severely ill adult COVID-19 patients only.
However, the sensitivity of the assay may not necessarily be very affected, since reports show similar
seroconversion rates for patients with mild and severe COVID-19 disease despite generally lower SARS-CoV-
2 antibody titres in the former group [6,7,13]. Conversely, studies on the dynamics and detection of SARS-
CoV-2 antibodies in children are lacking and requires urgent attention.
In conclusion, our findings show that in an ELISA format the sensitivity of detecting total SARS-CoV-2 RBD-
specific antibodies is higher than that of assays detecting spike-specific IgA or IgG only. It is important to note
that the presence of SARS-CoV-2-specific antibodies does not necessarily correspond to protection against
SARS-CoV-2 infection and disease. In order to define antibody-mediated protection, further investigation of
virus-specific antibody functions that include neutralization and Fc-mediated effector functionality are
needed. Sero-epidemiological investigations together with longitudinal studies on sequential samples taken
from SARS-CoV-2 patients are necessary to characterize the spread of the virus and the long term protection
of the antibodies measured.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 9 of 15
Acknowledgements
The research has been conducted using the Danish National Biobank resource, supported by the Novo
Nordisk Foundation, grant number 2010-11-12 and 2009-07-28.
Ethical statement
Exemption for review by the ethical committee system and informed consent was given by the Committee
on Biomedical Research Ethics - Capital Region in accordance with Danish law on assay development projects.
Conflicts of interest
The authors declare no competing interests.
References
[1] Chan JF-W, Yuan S, Kok K-H, To KK-W, Chu H, Yang J, et al. A familial cluster of pneumonia associated
with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster.
Lancet 2020;395:514–23. doi:10.1016/S0140-6736(20)30154-9.
[2] World Health Organization. Coronavirus disease 2019 (COVID-19) Situation Report – 78 2020.
[3] European Centre for Disease Prevention and Control. Novel coronavirus disease 2019 (COVID-19)
pandemic: increased transmission in the EU/EEA and the UK – sixth update. Rapid Risk Assess 2020.
[4] Service RF. The standard coronavirus test, if available, works well—but can new diagnostics help in
this pandemic? Sci Mag 2020.
[5] Zou L, Ruan F, Huang M, Liang L, Huang H, Hong Z, et al. SARS-CoV-2 Viral Load in Upper Respiratory
Specimens of Infected Patients. N Engl J Med 2020. doi:10.1056/NEJMc2001737.
[6] To KK, Tak O, Tsang Y, Leung W, Tam AR, Wu T, et al. Articles Temporal profiles of viral load in posterior
oropharyngeal saliva samples and serum antibody responses during infection by SARS-CoV-2 : an
observational cohort study. Lancet Infect Dis 2020;3099:1–10. doi:10.1016/S1473-3099(20)30196-1.
[7] Zhao J, Yuan Q, Wang H, Liu W, Liao X, Su Y, et al. Antibody responses to SARS-CoV-2 in patients of
novel coronavirus disease 2019. Clin Infect Dis (in press). doi:https://doi.org/10.1093/cid/ciaa344.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 10 of 15
[8] Guo L, Ren L, Yang S, Xiao M, Chang D, Yang F, et al. Profiling Early Humoral Response to Diagnose
Novel Coronavirus Disease (COVID-19). Clin Infect Dis (in press). doi:10.1093/cid/ciaa310.
[9] Bao L, Deng W, Gao H, Xiao C, Liu J, Xue J, et al. Reinfection could not occur in SARS-CoV-2 infected
rhesus macaques. bioRxiv 2020:2020.03.13.990226. doi:10.1101/2020.03.13.990226.
[10] Okba NMA, Muller MA, Li W, Wang C, GeurtsvanKessel CH, Corman VM, et al. SARS-CoV-2 specific
antibody responses in COVID-19 patients. medRxiv 2020:2020.03.18.20038059.
doi:10.1101/2020.03.18.20038059.
[11] Thevarajan I, Nguyen THO, Koutsakos M, Druce J, Caly L, van de Sandt CE, et al. Breadth of concomitant
immune responses prior to patient recovery: a case report of non-severe COVID-19. Nat Med 2020.
doi:10.1038/s41591-020-0819-2.
[12] Cassaniti I, Novazzi F, Giardina F, Salivaro F, Sachs M, Perlini S, et al. Performance of VivaDiagTM
COVID-19 IgM/IgG Rapid Test is inadequate for diagnosis of COVID-19 in acute patients referring to
emergency room department. J Med Virol 2020;n/a. doi:10.1002/jmv.25800.
[13] Long Q, Deng H, Chen J, Hu J, Liu B, Liao P, et al. Antibody responses to SARS-CoV-2 in COVID-19
patients: the perspective application of serological tests in clinical practice. medRxiv
2020:2020.03.18.20038018. doi:10.1101/2020.03.18.20038018.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 11 of 15
Tables
TABLE 1. Analytical sensitivities, specificities, and predictive values for SARS-CoV-2 antibody detection
Number (%) of serum samples
Assay Case sera testing
positive Control sera testing
negative
PPV (%) NPV (%)
ELISA
Wantai Total Ab 28/30 (93) 82/82(100) 28/28 (100) 82/84 (98)
Euroimmun IgAa 28/30 (93) 76/82 (93) 28/34 (82) 76/78 (97)
Euroimmun IgGa 20/30 (67) 79/82 (96) 20/23 (87) 79/89 (89)
Point-of-care test
Dynamiker 27/30 (90) 32/32 (100) 27/27 (100) 32/36 (89)
CTK Biotech 27/30 (90) 32/32 (100) 27/27 (100) 32/36 (89)
AutoBio Diagnostics 28/30 (93) 32/32 (100) 28/28 (100) 32/25 (91)
Artron Laboratories 25/30 (83) 17/17 (100) 25/25 (100) 17/23 (74)
Acro Biotech 4/5 (80)b 12/15 (80) 4/7 (57) 12/13 (92)
Alltest Biotech 1/1 (100)b 13/15 (87) Too few tested Too few tested
aBorderline data were considered negative.
bDue to comparatively poorer assay performance in an initial round of testing, further testing were suspended.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 12 of 15
TABLE 2. False positive results for sera from controls diagnosed with infections other than SARS-CoV-2
ELISA Point-of-Care Tests
Euroimmun IgA Euroimmun IgG
Virus-specific antibodies detected
# sera tested
Positive Borderline
Positive Borderline # sera
tested Acro
Biotech Alltest
Biotech
Unknown 10 1 1 2
hCoV-OC43 1 1
hCoV-HKU1 2 1 2 1 (IgM only)
hCoV-NL63 2 2
Adv 4 1 1 1 1
Infl A 5 0
Infl B 7 0
Infl A, Infl B 9 2 1
Infl A, Infl B, Adv 6 1 2 1 3 1 (IgM only) 1 (IgM only)
Infl A, Adv 4 1 1
Infl B, Adv 2 1
Infl A, Infl B, RSV 2 0
Infl A, Infl B, RSV, Adv 3 2
Adv, RSV 1 1 1
Infl A, RSV, Adv 1 1 1
Infl B, RSV 1 0
EBV 10 2 0
CMV 1 0
CMV, EBV 2 0
Dengue 9 1 1 2 1 (IgM only) 1 (IgM only)
Total 82 6 9 3 1 20 3 2
hCoV – human coronavirus; Adv – adenovirus; Infl – influenza; RSV – respiratory syncytial virus; EBV – Epstein-Barr virus; CMV – cytomegalovirus.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 13 of 15
Figures
Figure 1. SARS-CoV-2 antibody ELISA assay performance. A) ELISA data distribution obtained for control
and SARS-CoV-2 case sera using three commercial ELISA kits. Lines represents median values with
interquartile ranges. The dotted lines indicate the respective cut-off values recommended by manufacturer
to determine positive and negative test results. The grey zone marks the ratio range with borderline results.
B) ROC curves for the respective ELISAS assays.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 14 of 15
Figure 2. Analytical sensitivities of SARS-CoV-2 serologic assays in relation to the duration of illness: 7 to
13 days (n = 7), 14 to 20 days (n = 15), and ≥21 days (n = 8).
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint
Page 15 of 15
Figure 3. Agreement between ELISA and POC test results. A) Individual COVID-19 patient with PCR-
confirmed SARS-CoV-2 infection. Each row represents a patient, each column a serological test, a blue block
represents a positive result and a white block a negative result. B) The proportion of results that agreed
between two assays.
. CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)
The copyright holder for this preprint this version posted April 10, 2020. .https://doi.org/10.1101/2020.04.09.20056325doi: medRxiv preprint